In an image processing apparatus, an extracting unit extracts mutually the same region of interest from each of a plurality of pieces of three-dimensional image data corresponding to mutually-different time phases. Further, a position determining unit determines, on the basis of feature points included in the pieces of three-dimensional image data, a position used for superimposing together the regions of interest extracted by the extracting unit from the pieces of three-dimensional image data, in substantially the same position of a subject. After that, a display controlling unit changes a display format of each of the regions of interest extracted by the extracting unit from the pieces of three-dimensional image data so as to be mutually different and causes a superimposed image to be displayed by superimposing the regions of interest together in the position determined by the position determining unit.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. An image processing system comprising: processing circuitry configured to extract a mutually-same region of interest from each of a plurality of pieces of three-dimensional image data corresponding to mutually-different time phases; determine, on a basis of feature points included in the pieces of three-dimensional image data, a position used for superimposing together the regions of interest extracted from the pieces of three-dimensional image data, in a substantially same position of a subject; generate a plurality of parallax image groups that the pictures to be viewed change in accordance with shifting of the viewpoints of the viewer, by performing a rendering process on each of the pieces of three-dimensional image data while moving a point-of-view position by a predetermined parallactic angle, so as to include the regions of interest; hide a part of a region of interest to be on the outside so that a region of interest to be on the inside is exposed, when each of the regions of interest in the plurality of parallax image groups is superimposed; and display, on a display, a superimposed stereoscopic image formed by superimposing the plurality of parallax image groups, so as to superimpose the regions of interest together in the position determined.
An image processing system aligns and displays time-series 3D medical images to reveal changes. It extracts the same region of interest (ROI) from multiple 3D scans of the same subject taken at different times. Feature points within these scans are used to determine the best position to overlay the ROIs, ensuring they align anatomically. For each time point, the system generates parallax images by rendering the 3D data from slightly different viewpoints. Finally, the parallax images are superimposed to create a stereoscopic (3D) display. Overlapping regions are processed such that inner ROIs are visible, while outer ROIs are partially hidden, emphasizing changes over time in a combined 3D view.
2. The image processing system according to claim 1 , wherein the processing circuitry is configured to change opacity of each of the regions of interest extracted from the pieces of three-dimensional image data.
The image processing system described previously, which aligns and displays time-series 3D medical images to reveal changes by extracting the same region of interest (ROI) from multiple 3D scans, aligning them using feature points, generating parallax images, and superimposing them for a stereoscopic display, further modifies the opacity of each ROI. This allows users to see through overlaid regions to better understand the relative positions and changes over time of the different ROIs. For example, newer scans might be made more opaque to highlight recent changes, while older scans are more transparent.
3. The image processing system according to claim 1 , wherein the processing circuitry is configured to use a gravity point of each of the regions of interest extracted from the pieces of three-dimensional image data, as the feature points.
The image processing system described previously, which aligns and displays time-series 3D medical images to reveal changes by extracting the same region of interest (ROI) from multiple 3D scans, aligning them using feature points, generating parallax images, and superimposing them for a stereoscopic display, specifically uses the center of gravity of each ROI as a feature point for alignment. The system calculates the centroid of each ROI and uses these points to determine the transformation needed to align the ROIs across different time points.
4. An image processing apparatus comprising: processing circuitry configured to extract a mutually-same region of interest from each of a plurality of pieces of three-dimensional image data corresponding to mutually-different time phases; determine, on a basis of feature points included in the pieces of three-dimensional image data, a position used for superimposing together the regions of interest extracted from the pieces of three-dimensional image data, in a substantially same position of a subject; generate a plurality of parallax image groups that the pictures to be viewed change in accordance with shifting of the viewpoints of the viewer, by performing a rendering process on each of the pieces of three-dimensional image data while moving a point-of-view position by a predetermined parallactic angle, so as to include the regions of interest; hide a part of a region of interest to be on the outside so that a region of interest to be on the inside is exposed, when each of the regions of interest in the plurality of parallax image groups is superimposed; and display, on a display, a superimposed stereoscopic image formed by superimposing the plurality of parallax image groups, so as to superimpose the regions of interest together in the position determined.
An image processing apparatus aligns and displays time-series 3D medical images to reveal changes. It extracts the same region of interest (ROI) from multiple 3D scans of the same subject taken at different times. Feature points within these scans are used to determine the best position to overlay the ROIs, ensuring they align anatomically. For each time point, the system generates parallax images by rendering the 3D data from slightly different viewpoints. Finally, the parallax images are superimposed to create a stereoscopic (3D) display. Overlapping regions are processed such that inner ROIs are visible, while outer ROIs are partially hidden, emphasizing changes over time in a combined 3D view.
5. An image processing method comprising: extracting a mutually-same region of interest from each of a plurality of pieces of three-dimensional image data corresponding to mutually-different time phases; determining, on a basis of feature points included in the pieces of three-dimensional image data, a position used for superimposing together the regions of interest extracted from the pieces of three-dimensional image data, in a substantially same position of a subject; generating a plurality of parallax image groups that the pictures to be viewed change in accordance with shifting of the viewpoints of the viewer, by performing a rendering process on each of the pieces of three-dimensional image data while moving a point-of-view position by a predetermined parallactic angle, so as to include the regions of interest; hiding a part of a region of interest to be on the outside so that a region of interest to be on the inside is exposed, when each of the regions of interest in the plurality of parallax image groups is superimposed; and displaying, on a display, a superimposed stereoscopic image formed by superimposing the plurality of parallax image groups, so as to superimpose the regions of interest together in the determined position.
An image processing method aligns and displays time-series 3D medical images to reveal changes. It extracts the same region of interest (ROI) from multiple 3D scans of the same subject taken at different times. Feature points within these scans are used to determine the best position to overlay the ROIs, ensuring they align anatomically. For each time point, the method generates parallax images by rendering the 3D data from slightly different viewpoints. Finally, the parallax images are superimposed to create a stereoscopic (3D) display. Overlapping regions are processed such that inner ROIs are visible, while outer ROIs are partially hidden, emphasizing changes over time in a combined 3D view.
6. A medical image diagnosis apparatus comprising: processing circuitry configured to extract a mutually-same region of interest from each of a plurality of pieces of three-dimensional image data corresponding to mutually-different time phases; determine, on a basis of feature points included in the pieces of three-dimensional image data, a position used for superimposing together the regions of interest extracted from the pieces of three-dimensional image data, in a substantially same position of a subject; generate a plurality of parallax image groups that the pictures to be viewed change in accordance with shifting of the viewpoints of the viewer, by performing a rendering process on each of the pieces of three-dimensional image data while moving a point-of-view position by a predetermined parallactic angle, so as to include the regions of interest; hide a part of a region of interest to be on the outside so that a region of interest to be on the inside is exposed, when each of the regions of interest in the plurality of parallax image groups is superimposed; and display, on a display, a superimposed stereoscopic image formed by superimposing the plurality of parallax image groups, so as to superimpose the regions of interest together in the position determined.
A medical image diagnosis apparatus aligns and displays time-series 3D medical images to reveal changes to aid in diagnosis. It extracts the same region of interest (ROI) from multiple 3D scans of the same subject taken at different times. Feature points within these scans are used to determine the best position to overlay the ROIs, ensuring they align anatomically. For each time point, the system generates parallax images by rendering the 3D data from slightly different viewpoints. Finally, the parallax images are superimposed to create a stereoscopic (3D) display. Overlapping regions are processed such that inner ROIs are visible, while outer ROIs are partially hidden, emphasizing changes over time in a combined 3D view, facilitating the identification of clinically significant changes for diagnosis.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 17, 2014
March 21, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.